Hi all,
What IPv6 prefix lengths are people accepting in BGP from
peers/customers? My employer just got a /48 allocation from ARIN, and
we're trying to figure out how to support multiple end sites out of
this (probably around 10). I was thinking about assigning a /56 per
site, but looking at the BGP table stats on potaroo.net [1], it looks
like this is not too common (only .29% of prefixes). Thoughts?
Thanks,
--Richard
Are you talking about assigning /56s per POP, enterprise site?
If /56s are just in your iBGP..shouldn't be a problem. You're going to aggregate and just announce your /48 to your eBGP peers, yes??
If there are multiple indendent end sites then get a /48 for each of them.
A /48 is for a SITE not a ENTERPRISE.
If you are going to have each site connected separately to the outside world, you will want a /48 for each site.
If you are going to aggregate them internally, you can use whatever you want, although you should be able to get a /48 for each site anyway.
You don't want to announce anything longer than a /48 to the outside world.
-Randy
If you have to route them separately, your best bet is to go back to ARIN
under the Multiple Discreet Networks policy and get a block of /48s.
Tastes great, fewer problems.
Owen
Hi,
Hi all,
What IPv6 prefix lengths are people accepting in BGP from
peers/customers? My employer just got a /48 allocation from ARIN, and
we're trying to figure out how to support multiple end sites out of
this (probably around 10). I was thinking about assigning a /56 per
site, but looking at the BGP table stats on potaroo.net [1], it looks
like this is not too common (only .29% of prefixes). Thoughts?
Is it possible you should be using PA space from your ISP? An ISP would have
no issue providing a /48 per site. Not too many details were given about
your requirements or circumstances ... PA may fit.
if you have multiple sites you should request a direct assignmnet later
than /48. previous $employer recieved a /44 direct assignment on the
basis of north american footprint.
Richard's employer is exactly the kind of organization that has not
been able to effectively multi-home their discrete branch-offices on
the IPv4 Internet, because RIR allocation policy set the bar for
receiving IPv4 addresses for those small locations just high enough to
steer us away from that "feature" or "problem," depending on how you
look at it.
If every Richard Barnes announces a few dozen /48s into the global BGP
table, it will not be long before the 300k+ IPv4 BGP table looks neat
and organized, and the CIDR Report will come out each week with a
message begging e.g. Starbucks to aggregate their coffee shop wireless
hot-spots, instead of shaming Bell South for having a large number of
de-aggregates for their substantial ISP business.
Most people do not know about the "multi-homing feature" designed into
IPv6. Most people who do, seem to agree that it may not see enough
practical use to have meaningful impact on routing table growth, which
will no longer be kept in check by a limited pool of IP addresses and
policies that make it a little difficult for a very small network to
become multi-homed.
This may be another looming IPv6 headache without a sufficient
solution to set good practices now, before deployment sky-rockets.
Most people do not know about the "multi-homing feature" designed into
IPv6. Most people who do, seem to agree that it may not see enough
practical use to have meaningful impact on routing table growth, which
will no longer be kept in check by a limited pool of IP addresses and
policies that make it a little difficult for a very small network to
become multi-homed.This may be another looming IPv6 headache without a sufficient
solution to set good practices now, before deployment sky-rockets.
It's well known that IPv6 will require a scalable routing solution and that
one has not yet been developed. I'll be surprised if there isn't more
progress out of IETF on this issue in the near future.
Owen
Hi Richard,
Most major transit providers with IPv6 support filter at a maximum prefix length of /48 [1]
Also, some guidance in subnetting can be found at several places on the web [2]
[1] http://en.wikipedia.org/wiki/Comparison_of_IPv6_support_by_major_transit_providers
[2] http://en.wikipedia.org/wiki/IPv6_subnetting_reference
Dear Owen,
If you have some idea in your mind propose something for IETF. BoF sessions are open for completely new proposal: http://www.ietf.org/iesg/bof-procedures.html
or you can directly propose something for the particular wg:
http://datatracker.ietf.org/wg/
Best Regards,
Janos Mohacsi
Most people do not know about the "multi-homing feature" designed into
IPv6. Most people who do, seem to agree that it may not see enough
practical use to have meaningful impact on routing table growth, which
will no longer be kept in check by a limited pool of IP addresses and
policies that make it a little difficult for a very small network to
become multi-homed.This may be another looming IPv6 headache without a sufficient
solution to set good practices now, before deployment sky-rockets.It's well known that IPv6 will require a scalable routing solution and that
one has not yet been developed. I'll be surprised if there isn't more
progress out of IETF on this issue in the near future.
The RRG of the IRTF has spent the last two years on this topic. A summary of the discussed solutions can be find in:
A spin off of that activity is the LISP WG in the IETF (Locator/ID Separation Protocol (lisp))
Luigi